4 ms·

Some Unicode encodings require a char datatype that is 2+ bytes wide, hence a wide char.

• wolverine876 5y ago
  Right, so what does the GGP mean by "did not have to go to wide chars"?
  • dahfizz 5y ago
    Utf-8 can be implemented with plain old 8-bit chars. It does not need the wide chars like UCS-2, for example, where every character is 16 bytes. Therefore, by going to utf8 instead of ucs2, we did not have to go to wide chars.
  • cryptonector 5y ago
    "wide chars" == wchar_t, sizeof(char) == 1, sizeof(wchar_t) > sizeof(char).
  • masklinn 5y ago
    The code unit of UTF8 is the byte, so you only deal with bytes and byte sequences. Thus no issues of byte-order and such, which you have to deal with in UTF-16 and UTF-32 (as their code units are respectively 2 and 4 bytes).